1 - Getting Started ~5 mins
A good engineer does not stop at the first build. They test, learn, and try again.
Look at the bird feeder your group built earlier in this project. Which of our three jobs do you think it already does well: holding seed, keeping seed dry, or giving a stable perch? Hands up for your strongest guess.
Have every group's prototype ready on trays before the lesson starts. Do not lay out the full test kit yet; that appears in the test step.
Quick hands-up poll only: strongest criterion first. Do not run any tests here. The job of this beat is curiosity and a reminder that evidence, not hope, decides the next change.
Link to the project: pupils already named the user need and success criteria on their Design Brief earlier in this project. Point to those pages if groups have forgotten the three jobs.
Key question: How will we know if a change actually helped?
2 - How We Will Test Fairly ~7 mins
Every feeder gets the same checks: one level spoon of seed, a ten-second shower from the same height, and the same two fingers held on the perch for a slow count of three. That shared method is a fair test — same way every time so we can trust the comparison.
The checklist we score against is our success criteria: holds seed, stays dry, and stable perch. One careful change after the scores, then test again — that is iteration.
Score each job 0, 1 or 2:
- Holds seed: 0 = most seed falls out; 1 = holds some but leaks when moved gently; 2 = holds seed securely when lifted a little.
- Stays dry: 0 = seed soaked; 1 = partly wet; 2 = seed stays dry.
- Stable perch: 0 = tips or collapses under two fingers held for a slow count of three; 1 = wobbles but stays up; 2 = stays steady.
How does using the same spoon, shower and perch check keep the dry test fair?
Before this step: have the 0–2 rubric strip already written on the board or printed as a ready display strip. Do not spend lesson time writing the rubric out — the full wording is also on the IWB above so the class can see it while you talk. Leave that board strip up for both tests.
Project the short board version. Use the full table below as your reference while you talk. Keep this beat tight: name the shared method first, hang the three words on it, then run one quick scored demo.
| Concept | Why it matters | Example |
|---|---|---|
| Success criteria — the agreed checklist that shows whether a design does the job it was meant to do | Without shared criteria, groups argue about "good" and "bad" instead of looking at evidence | Holds seed, keeps seed dry, stable perch — the three jobs named on the Design Brief |
| Fair test of a prototype — testing every design the same way so comparisons can be trusted | If one group gets a gentle sprinkle and another gets a long pour, the dry result is about the test, not the feeder | Same seed amount, same ten-second shower from the same height, same two fingers held on the perch for a slow count of three |
| Iteration — one evidence-based change, then a re-test to see if the design improved | Real engineers improve from data, not from guessing several changes at once | Perch wobbled on the first test, so the group added one extra stick and tested stability again |
Worked model (live, about two minutes): Borrow one sturdy class demo feeder or a spare bottle mock-up. Think aloud through one full cycle once only:
- I wonder whether this feeder keeps seed dry.
- I predict the seed will stay mostly dry because the bottle top overhangs the tray.
- I test with one spoon of seed and the agreed ten-second pour, then look at the seed.
- I observed a few edge seeds damp, most dry — score 1 out of 2 for stays dry.
- I think a wider bottle-top roof would help — that would be the one change before a re-test.
Total possible first-test score: 6. Pupils record their own scores on the Investigation Journal page; the class later pools totals on the interactive.
Misconception to head off: changing three things at once and calling it an improvement. Iteration means one change linked to the weakest score, then a re-test.
Nature of STEM: this is how engineers work — evidence first, then a careful redesign, not a lucky rebuild.
3 - Keep the Shower Fair ~3 mins
Before any feeder is tested, we practise the shared ten-second control together.
Watch the countdown on the board. Listen for the start and the stop. Nobody pours water yet — this is only so everyone knows the rhythm we will use later at the pour stations.
Before this step
Open the fair-test-timer interactive on the IWB in countdown mode set to 10 seconds. Set up two pour stations (watering can or jug at each, each over a tray). Agree one pour height for both stations (for example the spout held at about 30 cm above the feeder, checked once with the metre stick).
How to run it
Practise one dry countdown with no feeder and no water so the class hears the rhythm. Explain: the timer is not a race; it is the control that keeps the test fair. The same ten-second pour will be used for every group's first dry test and again on the re-test.
Default to the shared ten-second shower so results can be pooled. Two stations mean two groups can shower at once during the test steps.
Look-fors: pupils can say why the same time and height matter; nobody reaches for a watering can in this practice beat.
4 - First Test of the Feeders ~14 mins
Test your feeder against all three success criteria. Finish the two table checks first, then join a pour-station queue for the shower. Score each check 0, 1 or 2 using the key on the board:
- Holds seed: 0 = most seed falls out; 1 = holds some but leaks when moved gently; 2 = holds seed securely when lifted a little.
- Stable perch: 0 = tips or collapses; 1 = wobbles but stays up; 2 = stays steady. Use a firm two-finger hold for a slow count of three — stronger than a real bird, so if the perch stays steady it is doing well.
- Stays dry: 0 = seed soaked; 1 = partly wet; 2 = seed stays dry.
1. Holds seed: put one level spoon of seed in the feeder, lift it a little, and watch.
2. Stable perch: hold two fingers down on the perch for a slow count of three.
3. Stays dry: set the feeder on the tray at a pour station and give it the ten-second shower from the agreed height.
Write the scores and what you saw on your Investigation Journal page before you change anything.
Setup
Stations or table groups with: the group's prototype, a tray, one spoon, a small pot of bird seed, paper towels. Two shared pour stations (watering can or jug + tray each) so two groups can run the stays-dry check at once. One adult keeps the pour queues calm and starts the ten-second countdown each time. Leave the 0–2 rubric strip on the board from step 2; the same descriptors are also on the IWB in this step.
Running the experiment
Groups run all three checks in the on-screen order: holds-seed and stable-perch first (dry hands at the table), then join a pour-station queue for stays-dry. They must record first-test scores before any rebuild. Circulate and press for observation language: what did you see? separate from what do you think it means?
Key questions while you circulate:
- Which criterion scored lowest, and what exactly did you observe?
- Did you keep the pour height and time the same as the class agreement?
- If two groups disagree on a score, what evidence would settle it?
Differentiation: less confident groups complete holds-seed and stable-perch first, then join the watering queue. Confident groups may add a short note on why a score was 1 rather than 2, still on the Investigation Journal page.
Safety: mop spills straight away; no running with wet floors; adult pours if a group is unsteady; wash hands after handling bird seed; no tasting seed.
If a feeder collapses: score what happened honestly, keep the pieces on the tray, and plan one repair as the improvement rather than a full rebuild.
5 - Pool the First-test Scores ~5 mins
Each group will call out its total first-test score out of 6. We will type the totals into the table on the board, then show the chart.
Which bars are tallest and shortest? What does a short bar tell an engineer to work on first?
Now a quick show of hands: which job scored lowest for your group — holds seed, stays dry, or stable perch? Which feeder job pulled most scores down across the class?
First-test feeder scores
Drive the data-recorder on the IWB in explore mode. Columns on screen are exactly Group and First-test total (out of 6). Add one row per group as totals are called (up to 12 rows).
After the table is filled, show the bar chart. Ask only what the chart can answer: tallest bars, shortest bars, and what a short bar means for the next change.
Then run a quick hands-up tally of weakest criterion (holds seed / stays dry / stable perch). Mark the three tallies on the board beside the chart so the class can see which job dragged scores down. Do not expect the totals chart alone to show criterion-level patterns.
Do not enter re-test scores yet; those come after the improvement. Pupils keep full criterion-by-criterion detail on the Investigation Journal page; the interactive holds the simple totals so the class can see the spread at a glance.
Prompts:
- What does a short bar tell an engineer to work on first?
- Why would it be unfair to compare totals if one group used a longer shower?
- Is a high score on two criteria enough if the third is zero?
Maths bridge: totals out of 6 are easy to compare; if useful, quickly find the class middle value of the first-test totals aloud without turning it into a long calculation lesson.
Unlock the full learning experience
You're previewing this lesson. Get full access to this lesson and hundreds more — each one ready to teach, with interactive activities, printable resources and pupil progress tracking built in.
This lesson is copyright of Coding Ireland 2017 - 2025. Unauthorised use, copying or distribution is not allowed.